Retaining or excising the supraspinatus tendon in complex proximal humeral fractures treated with reverse prosthesis: a biomechanical analysis in two different designs

Purpose We aimed to biomechanically evaluate the effect of the supraspinatus tendon on tuberosity stability using two different reverse shoulder arthroplasty (RSA) models for complex proximal humeral fractures (PHFs). Methods Four-part proximal humeral fractures were simulated in 20 cadaveric shoulders. Two different RSA designs were implemented: a glenosphere-medialized model and a glenosphere-lateralized model. Tuberosities were reconstructed, and displacement of bony fragments was measured (mm) by placing three sensors: in the humeral diaphysis (D), in the greater tuberosity (GT), and in the lesser tuberosity (LT). Axial forces were induced and measured in Newton (N). The test was performed twice in each specimen, with and without the supraspinatus tendon. The regression line (RL) was measured in mm/N. Results In the medialized model, the GT–D displacement was greater in the supraspinatus preserving model than that in the tendon excision model ( p < 0.001), as well as for the LT–D displacement ( p < 0.001). In the lateralized model, GT–D displacement and GT–LT distance were greater in the preserving model than that in the excision model ( p < 0.001, p = 0.04). Conclusion The supraspinatus tendon resection leads to a more biomechanically stable tuberosity construct when performing RSA for PHFs, while the rest of the rotator cuff tendons (infraspinatus and teres minor) are retained in the greater tuberosity. Level of evidence Basic science study. Cadaveric study.